摘要
实时系统是一种带有时间约束特性的复杂的计算机控制系统,其根本要求是计算机提供的计算能力必须能够满足各种各样的复杂实时任务的时间要求。随着企业信息系统的发展,多类型的硬实时、软实时与非实时任务共存于同一系统中的情况越来越广泛,具有这种复杂实时性要求的混合实时系统越来越受到研究者的关注,从而形成近年来以混合实时系统为对象的研究热潮。
混合实时消息系统通过对象请求代理,扩展了传统的C/S模式,建立了稳定、灵活、快速响应的三层分布式系统结构,解决了因底层服务设施暴露于应用而导致危险的问题,实现了客户远程调度服务;并基于JMS规范,实现统一的消息格式,提供多种通信模式,采用消息队列实现异步、可靠消息传输,消息事务和消息过滤功能,搭建了应用与应用之间,应用模块之间以及企业之间的消息交流通道。
混合实时消息的调度构建于一种层次调度模型的基础上,在调度对象上,将传统的简单任务扩充为开放式的复杂任务组-服务器对,并分析了系统对混合实时消息的可调度性;利用调度机制与调度策略的分离,解决了多类型调度策略的有机集成,实现了开放性的混合实时调度框架;并通过提出服务器虚拟化技术,采用面向服务的设计,以线程并行化为服务载体,完成混合实时消息的调度工作,保证了各类实时性要求。
最后,在地理信息协同标注环境下,测试并评估了混合实时消息系统的主要功能与性能,结果表明系统具有较好的适用价值和研究参考价值。
The real-time system is a kind of multi-task with time limit property complex controlling system by the computer. It’s one of the based requirements that the computer power offered by CPU must satisfy the time limit of all various kinds of complex real-time task to be scheduled. With the development of enterprise information system, the real-time system with complex real-time requirement composed with various tasks is focused on by the researcher in the real-time field, which recently causes to the hot research of mixed real-time message system.
The mixed real-time message system in the paper is an important part supported the Grdi-based Collaborative Geographic Information System (CoGIS), which is a sub-project of the National 10th Five-Plan. The system extends the traditional C/S model using the object request agent technology, and builds a stable and flexible 3-hiberarchy architecture with fast response to solve the danger caused by the exposure of inside infrastructure, and realize the remote call from clients. Based on the Java Message Specification (JMS), standardizes the union message format and supports various communication models. By message queue, the system can message with the property of, which realizes the communication among the enterprise applications and inside models in the application.
Based on the scheduling model of the mixed real-time message system, the scheduled object is expands to be the pair of task set and server. A scheduling model based on the virtual server preview gained the process time is proposed in the paper to satisfy with the requirement of the real-time cooperative system. The scheduling model is proved to be schedulable. By separating the scheduling mechanism and strategy, realizes the open schedule framework. And uses the multi-thread parallel technology, which makes it sure that several scheduler algorithms work together in a system.
Later, in the cooperative marking environment of geography information system, the scheduling effective is acceptable according to the model evaluating and test result. And the mixed real-time message system is valuable for using and researching.
引文
[1] Giuseppe Lipari, Sanjoy Baruah, A Hierarchical Extension to the Constant Bandwidth Server Framework, Proceedings of the Real-Time Technology and Applications Symposium. Taipei. Taiwan. May 2001. IEEE Computer Society Press. 3~13
[2] Liu C.L. and Layland J.W. , Scheduling Algorithms for Multiprogramming in a Hard real-time Environment, Journal of ACM 20(1), 1973. 46~61
[3] Kim, K. H. Ishida, Masaki, Juqiang Liu. An Efficient Middleware Architecture Supporting Time-Triggered Message-Triggered Objects and an NT-based Implementation. Proc. of the 2nd IEEE CS Int’l Symp. on Object-Oriented Real-time Distributed Computing, St. Malo, France, May, 1999. 166~178
[4] K. H. Kim, Juqiang Liu, Moon-Hae Kim. Deadline Handling in Real-Time Distributed Objects. Proc. IEEE CS 3rd Int’l Symp. on Object-oriented Real-time distributed Computing, Newport Beach, CA, March 2000. 7~15
[5] Deng Z, Liu JWS. Scheduling Real-Time Applications in Open Environment. In: Proceedings of the 18th IEEE Real-Time Systems Symposium. Los Alamitos, CA: IEEE Computer Society Press, 1997. 308~319
[6] 秦 啸, 庞丽萍, 韩宗芬, 李胜利. 分布式实时系统的容错调度算法. 计算机学报, 2000, 23(10): 1056~1063
[7] Lipari G, Baruah S. A hierarchical extension to the constant bandwidth server framework. In: Proceedings of the 7th IEEE Real Time Technology and Applications Symposium, Los Alamitos, 2001. 26~35
[8] Marco Spuri, Giorgio Buttazzo, Fabrizio Sensini. Robust Aperiodic Scheduling under Dynamic Priority Systems, Proceedings of the 16th IEEE Real-Time Systems Symposium, Italy, 1995. 210~219
[9] Kim, Liu Juqiang. QoS-driven Resource Management in Real-Time Object Based Distributed Computing System. Proc. The 8th IEEE Workshop on Future Trends of Distributed Computing Systems, Bologna, Italy, 2001. 212~221
[10] Bogdan Doytchinov, John Lehoczky, Steven Shreve. Real-Time Queues in Heavy Traffic with Earliest-Deadline-First Queue Discipline. The Annals of AppliedProbability, 2001, 11: 332~378
[11] S. Tai, Z. Deng, M. Shankarand, M. Storch. Probabilistic performance guarantee for real-time tasks with varying computation times. In Proceedings of IEEE Real-Time Technology and Application Symposium. 2003. 779~792
[12] Lin Chuang. Performance analysis of request dispatching and selecting in Web server clusters. Chinese Journal of Computers, 2000, 23(5): 500~508
[13] Marco Spuri, Giorgio C. Buttazzo, Efficient Aperiodic Service under Earliest Deadline Scheduling, 1994. 162~174
[14] Chen Lee, Ragunathan Rajkumar, Cliff Mercer, Experience with Processor Reservation and Dynamic QOS in Real-Time Mach, Proceedings of 23th Real-Time System Symposium, Bolin, 2001. 1011~1025
[15] Kane Kim, Juqiang Liu, Going Beyond Deadline-Driven Low-level Scheduling in Distributed Real-Time Computing Systems, Department of Electrical and Computer Engineering University of California
[16] Kim, K. H. Ishida, Masaki, Juqiang Liu, An Efficient Middleware Architecture Supporting Time-Triggered Message-Triggered Objects and an NT-based Implementation, Proc. 2nd IEEE CS Int’l Symp. on Object-Oriented Real-time Distributed Computing, St. Malo, France, 1999. 656~669
[17] Kim, Liu Juqiang., QoS-driven Resource Management in Real-Time Object Based Distributed Computing System, Proc. of the 8th IEEE Workshop on Future Trends of Distributed Computing Systems, Bologna, Italy, 2001. 945~954
[18] Pierre G. Jansen, Sape J. Mullender, Paul J.M.Havinga, Hans Scholten., Lightweight EDF Scheduling with Deadline Inheritance, Proc. of the 8th IEEE Workshop on Future Trends of Distributed Computing Systems, Bologna, Italy, 2001. 965~976
[19] 金海, 李其胜, 邹德清. 信息服务网格中资源代理的实现策略, 计算机软件与应用, 2004, 21(1) : 55~57
[20] 邹勇, 李明树, 王青. 开放式实时系统的调度理论与方法分析. 软件学报, 2003, 14(1): 83~90
[21] 庞丽萍, 黎时才, 陈汉华, 袁平鹏. 基于协同环境的混合实时任务调度模型.华中科技大学学报, 33(261): 121~124
[22] 宾雪莲, 杨玉海, 金士尧. 一种有限优先级的静态优先级分配算法. 软件学报, 2004, 15(6): 815~822
[23] 黄文广, 于士齐. 周期多帧任务的固定优先级调度算法的调度分析. 计算机研究与发展, 2001, 38(2): 240~245
[24] 秦啸, 韩宗芬, 庞丽萍, 李胜利. 混合型实时容错调度算法的设计和性能分析. 软件学报, 2000(5): 212~221
[25] Ray-I Chang, Wei-Kuan Shih, Ruei-Chuan Chang, Real-Time Disk Scheduling for Multimedia Applications with Deadline- Modification-Scan Schema, The International Journal of Time- Critical Computing Systems. 2001. 134~146
[26] Bogdan Doytchinov, John Lehoczky, Steven Shreve, Real-Time Queues in Heavy Traffic with Earliest-Deadline-First Queue Discipline, The Annals of Applied Probability, 2001, 11: 102~112
[27] Scheduling of Hard Read-Time Periodic Systems with Various Kinds of Deadline and Offset Constraints. Annals academic, 1999. 1251~1260
[28] Z. Deng, J.W.S.Liu, Scheduling Real-Time Applications in an Open Environment, Proc the 8th IEEE Workshop on Future Trends of Distributed Computing Systems, 1997. 421~432
[29] Chi-Sheng Shih, Jane W.S.Liu, State-Dependent Deadline Scheduling, Proc. of the 23rd IEEE Real-Time Systems Symposium. Italy, 2001. 63~72
[30] Damien Deville, Christophe Rippert, Gilles Grimaud.Trusted Collaborative Real Time Scheduling in a Smart Card Exokernel. INRIA, 2001. 321~333
[31] Stephen Brosky. Symmetric Multiprocessing and Real-time in PowerMAX OS.
[32] L. Abeni, G.Butazzo QoS guarantee using probabilistic deadlines.Pro. of the 11th Euromicro Conference on Real-Time Systems, 1999. 145~156
[33] M. Caccamo, G. Buttazzo. Elastic feedback control. In 12th Euromicro Conference on Real-time Systems IEEE, 2000. 421~433
[34] 进程调度实时化设计. 航天集团七七一研究所、西北工业大学计算机科学与工程系项目研制报告
[35] 黎宝华, 王姝暘, 庞丽萍. Linux 核心调度机制的实时性分析. 小型微型计算机系统, 1998, 19(2): 114~126
[36] 郭锐锋. 分布式实时操作系统中实时通信的研究. 小型微型计算机系统, 1995, 16(8): 35～40
[37] 王毅等. 分布式实时操作系统: 综合化系统资源管理机制的研究与设计. 计算机研究与发展, 1991(6): 32~44
[38] 穆兵. POSIX 实时标准介绍与应用. 计算机工程与设计, 1992(3): 66~78
[39] 王振宇. UNIX 实时化总体设计. 小型微型计算机系统, 1993(8): 101~112
[40] K. D. Lee, S. Kim. Traffic model and analysis for handoff performance in microcellular networks with directed retry. Proceedings in of the IEEE TENCON ’99, 1: 145~156
[41] T. S. Tai, Z. Deng, M. Shankarand.etc. Probabilistic performance guarantee for real-time tasks with varying computation times. In Proceedings of IEEE Real-Time Technology and Application Symposium. 2003. 97~108
[42] Buttazzo G., Sensini F. Robust Aperiodic Scheduling under Dynamic Priority Systems. Proc. of the 16th IEEE Real-Time Systems Symposium, 1995. 302~312
[43] T. M. Ghazalie, T. P. Baker. Scheduling Aperiodic Tasks in Dynamic Priority Systems. Real-Time Systems, 1995, 9(1): 46~59
[44] Mathai Joseph, Dynamic Priority Scheduling in Real-Time Systems: Specification, Verification and Analysis . Prentice Hall. 1995. 121~134
[45] 庞丽萍. 操作系统原理(第三版). 华中科技大学出版社, 2000
[46] 贝奇(美国). UNIX 操作系统设计. 北京大学出版社
[47] 袁由光等. 实时系统中的可靠性技术. 北京: 清华大学出版社, 1995
[48] 罗国庆. VxWorks 与嵌入式软件开发. 北京: 机械工业出版社, 2003
[49] 郑宗汉等. 实时系统软件基础. 北京: 清华大学出版社, 2003